Injected Light Emission of Silicon Carbide Crystals by K. LEHOVEC, C. A. Accardo, AND E. JAMGOCHIAN
InfoAge Homepage Back to the InfoAge HomepageBack Back to the Press Index Next Page
 

Camp Evans

Technical Publication

The Physical Review
Vol. 83, No. 3, 603-607
August 1, 1951 

 
 
Page 603 - 607
   evans logo

604                                                             LEHOVEC, ACCARDO, AND JAMGOCHIAN

 
Figure 2
 An explanation is proposed for the mechanism of the
yellow light emission.
 
II. EXPERIMENTAL ARRANGEMENT


     The experimental arrangement is shown diagramatically in Fig. 1. The current source for the crystal consisted of either a battery or a pulse generator. The silicon carbide crystal was silver-plated on one side and attached by silver paste to a brass plate mounted on a copper strip extending from a Kovar coolant container.  Two heating coils insulated by mica from the copper strip permitted measurements at elevated temperatures.  The light emission was studied under vacuum to prevent
water condensation at low temperatures. The light was
guided by a Lucite rod to an RCA 5819 photomultiplier
tube. The spectral response of the photomultiplier was
calibrated against a thermopile by using a prism
monochromator.  The relative sensitivity curve of the
Figure 3

 photomultiplier is shown in Fig. 2 (curve S). Its maximum
sensitivity was
Equation 0
for a wavelength of 4800A.  In Fig. 2 the transmissivities
(e .g., curves T1, T2) are shown for various Corning glass
filters. These were inserted between crystal and photomultiplier in order to determine the spectral distribution of the emitted light. The bell-shaped curves are the differences between the transmissivities of two successive filters multiplied by the photomultiplier sensitivity, e.g ., W12=S(T1-T2).
     The following procedure was used to calculate the
light intensity per unit wavelength from the photomultiplier
readings . Let the reading with filter 1 inserted be I1, and with a filter 2 be I2. The average light energy per unit wavelength in the range of W12 is then
Equation 1
(f is a numerical factor of the order of 10 resulting from
the geometry of the optical arrangement) . The appli-
Table 1
cation of Eq. (1) is justifiable if Ex does not change
much over the range of the integration. Ex is the spectral
density of the emitted light at the value of A which
halves the area of W12.  The values of X so chosen for
each of the bell-shaped curves are indicated by dotted
lines in Fig. 2.
     The silicon carbide crystals obtained through the
courtesy of the Carborundum Company were grown in
a commercial pile.  The crystals used were bluish in
color and had a surface oxide layer which was removed
by etching with hydrofluoric acids.8  Emitting areas were found by placing the crystal on abrass plate and probing
the surface with a point contact 22 volts negative with
respect to the plate.  The crystals passed currents of the
order of 10 ma at this voltage.  Suitable crystals were
silvered by evaporation and then mounted on the
brass plate with silver solder or silver paste.  In some
cases the base of the crystal was coated with zirconium
by heating zirconium hydride in contact with the
crystal in vacuum.  Silver soldering to the zirconium
layer is possible.9
---------
8 H. G. Heine and P. Scherrer, Helv. Phys . Acta 13, 489 (1940) .
9 C. S. Pearsall and T. B. Zingeser, Metal to Non-Metallic
Brazing  (Research Laboratory of Electronics, M.I .T., April 5,
1949), Technical Report 104 (unpublished) .

Page updated February 7, 2004   page created February 7, 2004

InfoAge Homepage Back to the InfoAge HomepageBack Back to the Press Index Next Page